Precision Farming and Health Monitoring in Agric-Workers using Wearable AI Devices

Abstract

This paper introduces a smart precision farming system with wearable health monitoring to optimise the productivity of agriculture and guarantee the safety of agri-workers. The study attempts to combat the weaknesses of conventional agricultural systems that tend to overlook real-time observation of physiological status of workers like stress and fatigue. This was designed in a layered Internet of Things architecture to allow real time transmission and analysis of continuous data over wearable sensors and environmental sources. The dataset used to develop a model was WESAD, and preprocessing, such as normalisation, filtering, and segmentation, were used to enhance the quality of the data and guarantee the high level of the extractions of the features. The model was an Artificial Neural Network (ANN) that was applied with the help of TensorFlow and Keras in order to distinguish between agri-worker conditions of stress, fatigue, and neutral. Results obtained from the independent test dataset demonstrated strong predictive capability, achieving 94.2% accuracy (95% CI: ±1.8%), 92.7% precision (±2.1%), 91.8% recall (±2.3%), and 92.2% F1‑score (±2.0%). Comparison with baseline models showed that the proposed ANN outperformed Support Vector Machine (SVM), Random Forest (RF), and Long Short‑Term Memory (LSTM) networks. Additionally, simulation across 100-time windows produced an overall system accuracy of 96% with an average response time of 1.2seconds, confirming real-time efficiency and consistency. The results show that the implementation of Artificial Intelligence alongside wearable sensors offers a dependable and scalable method of continuously tracking the wellbeing of agri-workers. The proposed system could help improve the early identification of stress‑related disorders, facilitate prompt intervention, and enable safer, more efficient, and sustainable agricultural methods. though field validation with actual agricultural workers remains necessary.